INDEX

Explanations

words related to negative emotions, such as disappointment, disgust, and disaster

oai_token-act-pair · gpt-3.5-turbo

words related to disappointment or negative sentiments

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kinnikuman

-0.76

akeru

-0.74

glers

-0.73

hetti

-0.70

 Juliet

-0.65

 Lans

-0.63

Reviewer

-0.61

Hod

-0.61

 Herm

-0.61

 Goth

-0.61

POSITIVE LOGITS

ruption

1.07

cipl

1.04

rup

1.03

placed

1.01

comfort

0.99

abled

0.98

puted

0.97

ciples

0.96

licted

0.95

rupt

0.95

Activations Density 0.007%

words related to negative emotions, such as disappointment, disgust, and disaster

words related to disappointment or negative sentiments

No Comments

No Known Activations